Re: [PATCH] mm: hugetlb: fix HVO crash on s390

[Joao Martins <joao.m.martins@oracle.com>]

On 28/10/2025 16:14, Heiko Carstens wrote:
> On Tue, Oct 28, 2025 at 04:05:45PM +0000, Joao Martins wrote:
>>> diff --git a/mm/hugetlb_vmemmap.c b/mm/hugetlb_vmemmap.c
>>> index ba0fb1b6a5a8..5819a3088850 100644
>>> --- a/mm/hugetlb_vmemmap.c
>>> +++ b/mm/hugetlb_vmemmap.c
>>> @@ -48,6 +48,15 @@ struct vmemmap_remap_walk {
>>>  	unsigned long		flags;
>>>  };
>>>  
>>> +static inline void vmemmap_flush_tlb_all(void)
>>> +{
>>> +#ifdef CONFIG_S390
>>> +	__tlb_flush_kernel();
>>> +#else
>>> +	flush_tlb_all();
>>> +#endif
>>> +}
>>> +
>>
>> Wouldn't a better fix be to implement flush_tlb_all() in
>> s390/include/asm/tlbflush.h since that aliases to __tlb_flush_kernel()?
> 
> The question is rather what is flush_tlb_all() supposed to flush? Is
> it supposed to flush only tlb entries corresponding to the kernel
> address space, or should it flush just everything?
> 
The latter i.e. everything

At least as far as I understand

> Within this context it looks like only tlb flushing for the kernel
> address space is required(?)

That's correct. We are changing the vmemmap which is in the kernel address
space, so that's the intent.

flush_tlb_all() however is the *closest* equivalent to this that's behind an
arch generic API i.e. flushing kernel address space on all CPUs TLBs. IIUC, x86
when doing flush_tlb_kernel_range with enough pages it switches to flush_tlb_all
(these days on modern AMDs it's even one instruction solely in the calling CPU).